An Insightful Theoretical Interpretation of Paper pH Surface Electrode

Article Cite This: J. Phys. Chem. C XXXX, XXX, XXX-XXX

pubs.acs.org/JPCC

An Insightful Theoretical Interpretation of Paper pH Surface Electrode Data Reveals a New Paradigm for the Definition of Paper Acidity Amerigo Beneduci,* Francesco Dalena, Bruna Clara De Simone, Ilaria Lania, Giovanni De Filpo, Maria Caterina Gallucci, and Giuseppe Chidichimo* Department of Chemistry and Chemical Technology, University of Calabria, Via P. Bucci, Cubo 15D, 87036 Arcavacata di Rende (CS), Italy ABSTRACT: Electrochemical measurement of pH on paper artifacts is a very important tool to assess the state of paper conservation. Different methods using different types of electrodes have been standardized to provide reproducible pH values of paper. One of these methodologies is not very invasive since it is made on the paper surface by means of a surface electrode in contact with the paper through a small layer of water previously added at the measurement point. This method is very convenient but suffers from the fact that the pH value given by the instrument is continuously changing during the measurement with a very fast rate in the first seconds and tending to reach a plateau value after very long time. Usually, the measure is taken at a time after which the rate of pH change is below a given value, i.e., when the value can be considered constant. This condition is usually signaled by an automatic instrumental alert provided to the operator. This pH value, however, is not the asymptotic one, which can be reached only at infinite time. Here we have investigated this behavior showing that it is simply due to the gradual diffusion of water away from the measurement region under the electrode. A theoretical model, which reproduces very well the time dependence discussed above, has been developed. The validity of the model has been completely proven by taking in consideration experiments where water diffusion is prevented in a simple confining apparatus. This work shows that it is possible to eliminate any factor affecting accuracy and reproducibility of the pH data for paper measured by the surface electrode method.

■

99).7 They differ with regard to the H3O+ extraction process from the paper. The first two methods consist of the extraction of hydrogen ions from a suspension of paper in water at 100 and 25 °C, respectively, for 1 h. The TAPPI contact method involves the use of a surface flat electrode. The above procedures have been recently compared and reviewed considering measurement repeatability, systematic error, and paper sample consumption.8 The contact method can be considered almost noninvasive, since pH measurement is simply made by placing the flat electrode surface on the top of a water droplet used for wetting the paper at the point where the measurement needs to be done. The H3O+ ion concentration generated by the reaction of water with the acid sites of the paper is then detected. The method appears to be very simple and convenient, but the investigator has to deal with the following data interpretation. During the measurement the instrument detects pH values which change quite rapidly immediately after the electrode contact with the wetted surface of the paper and keep growing,

INTRODUCTION One of the major problems with preservation of cultural heritage is the prevention of damage to paper and bookbinding in historic books and manuscripts that can be caused by the effect of aggressive atmospheres, humidity and temperature excursions, alterations of paper constituents, and added materials such as ink and hand coloring with pigments or dyes.1 Deterioration of paper-based materials is mainly due to the degradation of cellulose caused by a number of factors.2 One of the most feared is chemical attack due to acidic hydrolysis. For this reason, it is necessary to measure paper pH, because it is one of the most important parameters defining its longevity.3 An early scientific publication relating the measurement of paper pH with a glass electrode, in conjunction with a chloridecalomel reference electrode, is that of Herbert F. Launer (1939),4 who proposed the use of this method for pH measurements of hot extracts from paper, in contrast with the previously used methods of the total acidity.4 Nowadays there are three different standardized procedures for measurement of paper pH, that make use of pH meters. They are the cold extraction (TAPPI T 509 om-02),5 the hot extraction (TAPPI T 435 om-02),6 and the contact methods (TAPPI T 529 om© XXXX American Chemical Society

Received: August 14, 2017 Revised: October 21, 2017 Published: October 25, 2017 A

DOI: 10.1021/acs.jpcc.7b08099 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C with gradually decreasing speed, to reach a stable value in times on the order of some minutes. Some of the instruments inform the experimenter to not take into consideration the pH values measured during the fast changing phase, giving a message of instability for the measurement. Since the pH continues to change even after this warning disappears, the problem is to understand the meaning of the pH values gradually seen by the electrodes and which of them is the most suitable to represent the acidity of paper. This work takes into consideration the physical and chemical mechanisms which generate the continuous drift of pH measured by surface electrodes on paper surfaces. One of the first pieces of evidence that came out during this study was that the major physical factor affecting pH measurements is the diffusion of water on the paper surface during the measurement time. Experiments have been made in two different conditions: according to the standard TAPPI method, i.e., allowing water to diffuse, and in the absence of water diffusion, according to the setup described in the Experimental Section. As we will see, the time evolution of the measured pH is completely different in the two cases. Notably, the asymptotic values obtained in the two mentioned conditions differ by more than 0.5 pH unit. Moreover, the time evolution of pH, determined in suspension conditions according to the TAPPI T 509 om-02, cold extraction method, is analogous the that observed in the absence of water diffusion, if the dilution factor (i.e., the paper/ water w/v ratio) is taken into account in both the experiments. A theoretical kinetic model has been developed, to fit the experimental data obtained both in the presence and absence of water diffusion, taking into account the chemical production of H3O+ ions and water diffusion.

Figure 1. Setup of the pH measurement experiments in standard conditions.

■

EXPERIMENTAL SECTION Experiments have been performed on a sacrificial acid book: a copy of conference book reports “XI Congresso Nazionale di Chimica Inorganica”. The pH has been measured by a HANNA- Leather&Paper HI 99171 pH meter with a surface glass electrode. Standard pH Measurements by Surface Electrode. Measurements of pH in standard condition have been made by placing a polypropylene plastic sheet beneath the investigated page. On each of the measurement points a small drop of MilliQ water (HANNA HI70960 conductive electrolyte solution) has been dripped by the droplet dosing cup bottle. The surface electrode has been then set on the water droplet. Both these operations have been made in rapid succession in a time lower than 1 s. The experimental setup is illustrated in Figure 1. pH values have been registered at intervals of 5 s, even during a first time interval of about 30 s when the instrument was giving an indication that the measure was unstable. Actually, no discontinuity has been observed in the trend of the gradually increasing pH values when the instrument signaled that the measure did enter in the stability regime. Measurements Made by Surface Electrode in Absence of Water Lateral Diffusion. In order to avoid water lateral diffusion during surface electrode pH measurements, discs of paper having the same diameter of the surface electrode have been used. Each disc has been inserted in a plastic cylinder having a perfectly flat bottom and a diameter of 0.1 mm larger than that of the disc (and of the electrode). As in standard measurements, a small droplet of water has been added to the paper, and measures have been taken by inserting the electrode in a plastic cylinder, as is shown in Figure 2.

Figure 2. Setup used to perform pH measurements in absence of water lateral diffusion. (A) Plastic cylinder and paper disc sample. (B) Electrode insertion into the cylinder to contact the disc and surface water.

Measurements Made by an Immersion Electrode (TAPPI Cold Extraction TAPPI T 509 om-02). In this case, 1.5 g of paper taken by pages 9 and 12 of the same book used in the experiments described above has been placed in a small beaker. A pH meter HANNA HI223 Calibration Check Microprocessor pH Meter, working with an immersion glass electrode, has been positioned on the top of the paper leaving a space of about 0.1 mm between the paper and the bottom of the electrode. A 10 mL portion of distilled water has been then added, and the pH has been measured.

■

RESULTS AND DISCUSSION The pH values measured according to the standard TAPPI 529 T om-99 procedure on three different points of the same sheet of the sacrificial book, are shown in Figure 3a. It can be noticed that, in every case, the trend of data does not show any discontinuity, despite the fact that the instrument signaled an instability of the measurements in the first 30 seconds after the electrode contact, while indicating a stable regime after this first half minute interval. The only change that can be noticed between these two different regimes is the slope of the data, which goes from a faster growing trend to a slower growing one. The question regarding the factors influencing the data evolution remains open. A completely different trend has been observed in the case of the measurements made on plastic confined discs (Figure 3b), B


Article

The Journal of Physical Chemistry C

Figure 4. Variation in time of pH data, measured by an immersion glass electrode. Data are taken from paper, from pages 9 and 12 of the sacrificial book (see the text). The weight ratio between the paper and the water used in the experiment was 0.15.

■

THEORETICAL MODEL According to the standard measurements of pH on the surface of book pages, the addition of a small water droplet on the measuring point is required. This droplet flattens under the electrode surface and then starts to diffuse laterally with respect to the electrode expanding the wetted surface in a gradually growing circular region. We hypothesize the occurrence of the following reaction in the wetted regions: Figure 3. Time variation of pH data measured by a surface glass electrode: (a) according to standard TAPPI T 509m procedure on different areas of page 24 (black and red points) and of page 187 (blue, dark cyan, magenta, points); (b) by preventing water diffusion using the experimental setup described in the Experimental Section. Black, red, and blue data points were taken from paper discs cut out from pages 7, 11, and 23 of the sacrificial book (see text) and confined in the bottom of a plastic cylinder to avoid water lateral diffusion. The paper-to-water weight ratio was 0.60.

k1

SH+ + H 2O ⇄ S + H3O+

(1)

k2

+

Here, SH indicates the acid sites of the paper that are able to generate H3O+ ions, and S represents the conjugate basic sites obtained after hydrogen ion extraction. k1 and k2 are the kinetic constants of the direct and inverse reactions. We will call K= k1/k2 the equilibrium constant of reaction 1. We will also hypothesize that the concentration of SH+ and S sites, present on the paper, is much larger with respect to the concentration of H3O+ ions generated by reaction 1. This hypothesis seems reasonable with the consideration that pH values are usually greater than 4, but we are aware that its full justification can be given only when considering the success of the model to reproduce the experimental data. We will show that this is actually the case. [SH+] and [S] can be then considered constant, during the evolution of reaction 1, that is to say

where water is not allowed to diffuse laterally with respect to the measuring electrode. Indeed, after a fast initial decay, the pH reaches an almost stable value within a time of the order of 30 s. Data taken by the immersion glass electrode, according to the TAPPI T 509 om-02 cold extraction method, are reported in Figure 4. It can be seen that the pH temporal trend is quite similar to that observed in Figure 3b. This is not surprising because diffusion of water through the paper cannot have a role on the time evolution of data, with the paper fully immersed in water since the start of the experiment. The limit values of the time evolution of pH data, of course, show some substantial differences with respect to the analogous limits obtained in the experiments reported in Figure 3b, but we will see in the next section that this effect is fully justified by taking into account the different paper/water w/v ratio (dilution factor). In order to explain the trend of data obtained in the two set of experiments described above, we developed a theoretical model which takes into consideration both the extraction reaction of H3O+ ions by the added water and the lateral diffusion of water that occurs when standard measurement are performed.

[SH+] = sa

(2)

[S] = s b

(3)

Taking into consideration reaction 1, the variation in time of the H3O+ concentration can be written according to the following kinetic equation: d [H3O+] = k1sa[H 2O(t )] − k 2s b[H3O+(t )] dt

(4)

+

The time variation of the H3O concentration depends on the variation of the water concentration in the region of the electrode surface. Even if the concentration of water consumed through reaction 1 is surely negligible, we cannot neglect the variation in the water concentration induced by the lateral diffusion of water with respect to the electrode surface. C


Article

The Journal of Physical Chemistry C The time variation of the water concentration due to diffusion can be calculated according to the first Fick’s law: ∂ [H 2O( r ⃗ , t )] = −D∇2 [H 2O( r ⃗ , t )] ∂t

∫0 ∫0

(5)

−

Here, D is the diffusion coefficient of water in the paper, when the variation of water concentration due to reaction 1 is neglected. This further approximation is fully justified due to the fact that the water concentration remains always much higher than that of the H3O+ ions. This becomes evident by performing some simple calculation. The moles of water added on the paper are around 10−3. The volume of paper in which this water is absorbed is on the order of 10−4 liter. The concentration of water in the wetted paper is that on the order of 10. There is a difference of more than 5 orders of magnitude between the water concentration and that of the H3O+ ions. Equation 5 can be solved in two dimensions, since water diffusion along the direction parallel to the electrode axis is prevented by the plastic sheet beneath the paper. The ∇2 operator can be conveniently expressed in cylindrical coordinates since both the water droplet placed on the paper beneath the electrode surface and the wetted surface (which gradually enlarges outside from the electrode boundary, during the measuring time) have cylindrical symmetry. ∇2 =

1∂ ⎛⎜ r ∂ ⎞⎟ 1∂ 2 ∂2 + 2 2 + 2 r ∂r ⎝ ∂r ⎠ r ∂θ ∂z

2π

∫0 ∫0 ∫0

R

L

= −D

∫0

R

R

∂ ∂ [H 2O( r ⃗ , t )] dr + r [H 2O( r ⃗ , t )] + ∂r ∂r 0

⎛∂ ⎞ ∂ [H 2O( r ⃗ , t )] dr = R ⎜ [H 2O( r ⃗ , t )]⎟ ⎝ ∂r ⎠R ∂r

(10)

Equation 10 relates the time derivative of the water concentration in the electrode region with the spatial gradient of the water concentration existing at the electrode contact edge. In order to solve this equation, a specific function for the edge gradient of the water concentration needs to be assumed. We are guided in this assumption by the following considerations. The edge gradient has its maximum value at the time t = 0 (time at which the electrode is placed on the water droplet) and reduces to zero for t = ∞ (time at which the diffusion process would reduce to zero in absence of other physical phenomena). The exponential function reported below has been chosen as a trial function since it has a minimum number of adjustable parameters:

(6)

⎛∂ ⎞ ⎜ [H O( r ⃗ , t )]⎟ = ϕ0e−αt ⎝ ∂r 2 ⎠R

On the basis of this assumption eq 10 can be rewritten as 2D −αt ∂ [H 2O(t )] = − ϕe R 0 ∂t

(7)

to obtain, finally [H 2O(t )] = [H 2O(0)] −

2Dϕ0 αR

(1 − e−αt )

(11)

We have now a specific function of the time for the concentration of water under the measuring electrode. This function can be inserted into eq 4 to calculate the time dependence of the H3O+ ion concentration in the electrode region:

(8)

2k1saDϕ0 d[H3O+(t )] (1 − e−αt )+ = k1sa[H 2O(0)] − dt αR

Here, L is the thickness of the page, and R is the electrode radius. The first member in eq 8 represents the time derivative of the total quantity of water (H2O)T(t) contained in the volume of paper beneath the electrode surface. Once the integration on the coordinates θ and z is performed in the second member of eq 8, the following result is obtained: R

R

∂2 [H 2O( r ⃗ , t )] dr = ∂r 2

∂ [H 2O( r ⃗ , t )] dr + ∂r

⎞ ∂ 2D ⎛⎜ ∂ [H 2O(t )] = − [H 2O( r ⃗ , t )]⎟ ⎠R ∂t R ⎝ ∂r

∫0 ∫ ∫

∫0

∫0

r

R

If the two members of the above equation are divided by the volume of the paper lying beneath the electrode (V = πR2L), the time derivative of the water concentration in the region of the glass electrode is obtained:

∂ [H 2O( r ⃗ , t )] dτ ∂t 2π R 1∂ ⎡ ∂ ⎤ ⎢r [H 2O( r ⃗ , t )]⎥⎦r dr dθ dz 0 0 r ∂r ⎣ ∂r

∂ (H 2O)T (t ) = − 2πLD ∂t

R

∫0

⎛∂ ⎞ ∂ (H 2O)T (t ) = − 2πLDR ⎜ [H 2O( r ⃗ , t )]⎟ ⎝ ∂r ⎠R ∂t

In order to calculate the amount of water in the wetted region under the measuring electrode, eq 7 can be integrated on the volume of this region: L

⎤ ∂⎡ ∂ ⎢r [H 2O( r ⃗ , t )]⎥⎦ dr = ∂r ⎣ ∂r

Then, we obtain

r, θ, and z are the cylindrical coordinates according to their usual meaning. The water concentration can be taken as a constant along the paper thickness due to the fact that this thickness is lower than 0.1 mm. Not only does the wetting process of paper maintain a cylindrical symmetry, but also we can further assume that it leads immediately to a uniform distribution of the water concentration along the z coordinate. Then, eq 6 can be simplified as ⎤ ∂ 1∂ ⎡ ∂ [H 2O( r ⃗ , t )] = −D ⎢r [H 2O( r ⃗ , t )]⎥⎦ ⎣ r ∂r ∂r ∂t

R

−k 2s b[H3O+(t )] (12)

The mathematical form of eq 12 suggests that its solution must be of the type:

⎤ ∂⎡ ∂ ⎢⎣r [H 2O( r ⃗ , t )]⎥⎦ dr ∂r ∂r

[H3O+(t )] = γ + β e−αt (9)

(13)

When eq 13 is substituted into eq 12 and the time independent terms are collected on the left, while the time dependent terms are collected at the second member, one obtains

Integration at the second member of eq 9 can be easily done as shown below: D


Article

The Journal of Physical Chemistry C ⎡ 2k1saDϕ0 ⎤ −αt ⎢k 2s bβ − αβ − ⎥e = k1sa H 2O(0)+ ⎣ αR ⎦ −

2k1saDϕ0 αR

− k 2s bγ

and eq 17 is obtained by substituting in eq 19, the value of H2O(t) given by eq 11. Equation 17b predicts that, due to the diffusion of water across the paper, the pH value measured by a glass surface electrode continuously increases in time to reach an asymptotic value, [H3O+(∞)], given by eq 18. In order to derive an equation that is useful to reproduce the data obtained in absence of water lateral diffusion, we can directly integrate eq 4. In this case, the time evolution of pH corresponds to the gradual proceeding of reaction 1 to its equilibrium state. The solution of eq 4 can be simplified by introducing a sound approximation, i.e., that, at any time, [H2O(t)] is always much larger than [H3O+(t)]. This allows the substitution of [H2O(t)] with [H2O(0)]. Equation 4 can then be written as

(14)

The conditions for which eq 14 holds at any time are that both the right member and the pre-exponential term on the left must be equal to zero. Under these conditions β and γ are easily calculated:

β=

γ=

2k1saDϕ0 αR(k 2s b − α)

(15a)

2Dϕ0 ⎫ k1sa ⎧ ⎨[H 2O(0)] − ⎬ k 2s b ⎩ αR ⎭

(15b)

d [H3O+] = c1 − c 2[H3O+(t )] dt

Then, we have the solution for the time dependence of the H3O+ ions in the electrode region: [H 2O+(t )] = +

c1 = k1sa[H 2O(0)]

αR(k 2sb − α)

pH = −log{[H3O+(0)]e−c2t + [H3O+(∞)](1 − e−c2t )} (21)

e−at

where, in this case

(16)

This solution does not hold when water lateral diffusion is absent, a condition for which the ratio D/α becomes fully undetermined with both D and α equal to zero. Equation 16 can be further simplified, thus reducing the number of independent adjustable parameters in order to fit the experimental data, if we soundly assume that the reaction kinetics of protons extraction is faster than water diffusion. In this condition, the term k2sb would be much greater than α, and eq 16 becomes [H3O+(t )] =

[H3O+(∞)] =

■

FITTING OF THE EXPERIMENTAL DATA Surface pH Data in the Presence of Water Diffusion (TAPPI T 529m Method). The experimental data reported in Figure 3a have been fitted by eq 17b, by varying the three fitting parameters pH(∞), pH(0), and α. The continuous lines in Figure 3 show the interpolation of the experimental data points while the obtained fitted parameters are reported in Table 1. The first comment concerns the important difference between the values of pH(0) and pH(∞) which is in the range 0.3−0.5 pH units. In order to ensure reproducibility on pH data measured by the surface electrode method, it is not appropriate to consider values measured at times where the detected pH continues to evolve in time, since we cannot give any precise

(17a)

or, in terms of pH pH = −log{[H3O+](0) − [H3O+](∞)e−αt }

(17b)

where ⎧ks ⎫ pH(0) = − log[H3O+(0)] = −log⎨ 1 a [H 2O(0)]⎬ ⎭ ⎩ k 2s b ⎪

⎪

is the acidity under the electrode surface at time zero, and pH(∞), the asymptotic acidity value remaining under the electrode, is given by

Table 1. Values of the Parameters pH(∞), pH(0), and α Obtained by Fitting the Data of Figure 3a with Equation 17ba

pH(∞) = −log[H3O+(∞)] = +

sample

(18)

black red blue dark cyan magenta

Equation 17 can be more straightforwardly obtained from eq 4 if it is assumed that reaction 1 reaches a stationary state in a very short time (seconds) such that d[H3O+]/dt = 0. In this case eq 4 becomes [H3O+(t )] =

k1sa [H 2O(t )] k 2s b

k1sa [H 2O(0)] k 2s b

while [H3O+(0)] is the pH of pure water. It is interesting to notice that the asymptotic pH value obtained in the case of blocked diffusion is equal to the value of pH extrapolated at time zero when using eq 17. This is not surprising when considering that eq 17 is strictly valid if reaction 1 reaches the equilibrium in a time which is negligible with respect to the time of water diffusion.

2Dϕ0(1 − e−αt ) ⎤ k1sa ⎡ ⎢[H 2O(0)] − ⎥ ⎥⎦ αR k 2s b ⎢⎣

⎧ 2k1saDϕ0 ⎫ ⎬ −log⎨[H3O+(0)] − k 2s bαR ⎭ ⎩

(20)

Integration of eq 20 leads to the following equation for the variation of pH with time:

2Dφ0 ⎫ k1sa ⎧ ⎨[H 2O(0)] − ⎬+ αR ⎭ k 2s b ⎩

2k1saDφ0

c 2 = k 2s b

pH(∞) 5.31 5.28 5.13 5.14 5.14

(0.01) (0.01) (0.01) (0.01) (0.01)

pH(0) 4.79 4.75 4.82 4.88 4.85

(0.02) (0.04) (0.03) (0.04) (0.03)

α (s−1) 0.046 (0.001) 0.043 (0.001) 0.0180 (0.0007) 0.018 (0.001) 0.024 (0.001)

a

Color according to the curves in Figure 3a. Values in parentheses are the standard deviations of the data.

(19) E


Article


Table 2. Values of the Parameters pH(∞), pH(0), and c2 Obtained by Fitting the Data of Figure 3b with Equation 20

meaning to these measurements. In addition, the use of the asymptotic pH value, pH(∞), does not seem to be the right choice since it depends on the value of the diffusion coefficient of water on the paper surface (eq 18). In our opinion, the most appropriate pH value that should be used to define the paper acidity is the extrapolated one at time zero, i.e., pH(0). This value depends only on the equilibrium of the reaction between the acid sites of the paper with the water contacting the electrode surface, which generates the hydrogen ions. It is worth mentioning that the difference between [H3O+(0)] and [H3O+(∞)] is proportional to the diffusion coefficient of water on the paper surface. This physical parameter is expected to be strongly dependent on paper porosity which, in turn, is expected to depend on the paper’s structural degradation. Paper aging usually leads to an increase of the number of pores (microholes) and thus to an increase of the water diffusion coefficient on paper. Our model allows for the knowledge of the exponential time constant α, a value that is expected to increase with an increase of the diffusion coefficient and, consequently, with an increase of the paper degradation. Therefore, we would expect that a useful correlation between this parameter and the conservation state of paper will be discovered, opening the way for new applications of the pH investigation by surface electrodes to study the conservation of paper historical artifacts. In light of the above considerations, we have a clear interpretation of the quite different trend of black and red curves measured on two points of page 24, with respect to that of blue, dark cyan, and magenta curves related to measurements taken from page 187. Table 1 shows that the diversification between the two groups of data is not linked to their different acidities with the values of [H3O+(0)] and [H3O+(∞)] being almost equal within the experimental error. We would like to underline that single measurements taken at time intervals ranging from 0 s up to 200 s would have been interpreted in terms of a different acidity of page 24 with respect to that of page 187, without taking into consideration the real mechanism of the time evolution of the experimental data, according to the model presented in this paper. The value of α is 2 times larger for page 24 than for page 187. This means that water diffuses faster on page 24 than on page 187, thus indicating that the paper has a different porosity for these two pages of the investigated book. Surface pH Data in the Absence of Water Diffusion. The experimental data taken in the absence of water diffusion (Figure 3b) confirms the physical model presented above. In this case, the acidity increases gradually with time to reach an asymptotic value. The trend is opposite to the one observed in the presence of water diffusion, due to the fact that only the gradual extraction of H3O+ by the water trapped in the region underlying the electrode is occurring. The parameters fitted by eq 20 are reported in Table 2. It can be seen that the values of pH(∞) obtained in the absence of diffusion (Table 2) are very similar to the pH(0) values obtained when the diffusion mechanism is active (Table 1), as is foreseen by the model developed in this paper. Suspension pH Data (TAPPI T 509 om-02 Cold Extraction Method). Finally, we comment on the results obtained by fitting the data reported in Figure 4, measured according to the cold extraction method, when the paper/water weight ratio was 0.15. In this case, the value of pH(0) is practically that of the pure Milli-Q water, while the value of [H3O+(∞)] is almost 0.6 unit higher than that measured in the

samplea black red blue suspensionb

pH(∞) 4.79 4.89 4.94 5.55

(0.01) (0.01) (0.01) (0.01)

pH(0) 5.25 5.74 5.30 6.89

(0.06) (0.01) (0.06) (0.02)

c2 (s−1) 0.063 (0.006) 0.026 (0.007) 0.056 (0.004) 0.0020 (0.0003)

a

Color according to the curves in Figure 3b. bData from Figure 4. Values in parentheses are the standard deviations of the data.

experiment where the diffusion of water has been artificially prevented. This difference is simply due to the 4 times higher paper/water weight ratio used in the suspension method with respect to that used in the surface method. This corresponds to a dilution factor of 4 and accounts for the observed difference in the final pH of 0.6 unit.

■

CONCLUSIONS This work has been motivated by the need to find an explanation of the continuous drifting of paper pH during the surface electrode measurement. A theoretical model which takes into account the diffusion of water through the paper under the electrode contact zone has been developed. This model has been tested by comparing experimental data taken according to the TAPPI T 529 om-99 procedure with data obtained when lateral water diffusion is prevented by confining the measurement zone in an appropriate way. The excellent agreement between the experimental time evolution of data and those calculated by means of the model leads to a clear identification of the dominant mechanisms influencing pH measurement. One of the identified mechanisms corresponds to the time evolution of the hydrogen ion concentration produced by the reaction between water and the acid sites of paper. The other mechanism is the one related to the time variation of the water concentration due to the diffusion of water molecules from the region lying beneath the surface electrode to surrounding regions of the paper. Of course, the two mechanisms are interconnected, but it has been shown that, with a good approximation, they can be decoupled, and very simple analytical solutions of the problems can be found. The model has been confirmed by fitting pH data obtained both in standard conditions and water diffusion prevented conditions. This paper finally suggests that, in order to have surface electrodes, and reproducible pH measurements on paper, the time evolution of the pH needs to be detected for sufficiently long time intervals and the extrapolated value at time zero needs to be calculated. We suggest the use of this value as the correct paper pH. In practical applications, therefore, we do not suggest the use of the water diffusion prevented method, because we are aware that the need to cut the paper artifact would cause an irreversible damage of that artifact. Instead, this invasive experiment has been merely used to show that, by using the TAPPI T 509 om procedure, it is possible to treat experimental data measured as a function of time to extract pH values not affected by diffusion. It would also be quite straightforward to implement a simple program into the pH-meter software able to automatically compute this extrapolated value after a reasonable time course pH measurement. Furthermore, we suggest that the interpolation of data time evolution allows the knowledge of parameters that could be correlated to the conservation state of the paper. F


Article


■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Amerigo Beneduci: 0000-0003-1185-9470 Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors are gratefully to the Italian Minister of University and Scientific Research (MIUR) for financial support under the project IT@CHA PON01_00625.

■

REFERENCES

(1) Faubel, W.; Staub, S.; Simon, R.; Heissler, S.; Pataki, A.; Banik, G. Non-Destructive Analysis for the Investigation of Decomposition Phenomena of Historical Manuscripts and Prints. Spectrochim. Acta, Part B 2007, 62, 669−676. (2) Abdel-Maksoud, G. Analytical Techniques Used for the Evaluation of a 19th Century Quranic Manuscript Conditions. Measurement 2011, 44, 1606−1617. (3) Strlič, M.; Pihlar, B.; Mauko, L.; Kolar, J.; Hočevar, S.; Ogorevc, B. A New Electrode for Micro-Determination of Paper pH. Restaurator 2005, 26, 159−171. (4) Launer, H. F. Determination of the pH Value of Paper. Part of J. Res. Natl. Bur. Stand. 1939, 22, 553−564. (5) Hydrogen Ion Concentration (pH) of Paper Extracts (Cold Extraction Method); TAPPI Test Method 2002 T509 om-02. (6) Hydrogen Ion Concentration (pH) of Paper Extracts (Hot Extraction Method); TAPPI Test Method 2002 T435 om-02. (7) Surface pH Measurement of Paper; TAPPI Test Method 1999 T529 om-99. (8) Strlic, M.; Kolar, J.; Koéar, D.; Drovsek, T.; Šelih, V. S.; Susiè, R.; Pihlar, B. What Is the pH of Alkaline Paper? e-Preserv. Sci. 2004, 1, 35− 47.

G


An Insightful Theoretical Interpretation of Paper pH Surface Electrode

Recommend Documents